RoonAppliance memory leak causing high RAM usage on Ubuntu 24.04.4 LTS (ref#FIB170)

Support
1

Hi! What’s not quite right with Roon?

· None of the above quite fits

None of the above quite fits

· None of these quite match

Tell us what's going on

· RoonAppliance memory leak — grows to ~60% of system RAM over a few days, repeats regularly

Running RoonServer 2.71 (build 1683), production, on Ubuntu 24.04.4 LTS (kernel 6.17.0-35-generic), 31GB RAM, x86_64.

RoonAppliance (the appliance process, not RoonServer itself) grows unbounded over a few days of uptime until it exhausts system memory. Most recent instance: after 3 days 20 hours uptime, RSS reached ~19GB (59.7% of total system RAM), pushing the host's 8GB swap to fully exhausted and RAM to 379MB free. Caught it manually before an OOM kill actually happened — journalctl -k shows no OOM events, but the system was at the edge.

This has recurred a few times. A systemctl restart roonserver.service reliably fixes it — RSS drops from 19GB to ~1.1GB immediately after restart — but it comes back over the following days.

No crash reports or errors in the Roon logs I could find pointing at a cause. Happy to grab diagnostics (Roon's built-in diagnostic export, specific log files) if that helps triage.

Tell us about your home network

· Wired network via over TP-Link TL-SG108PE Gigabit PoE Switch

2

Hey @Helge_Gudmundsen, welcome to the community, and thanks for the detailed report.

The memory growth you’re seeing is not expected, especially with the swap getting exhausted over a few days. Since a simple service restart clears it, we’d like to collect a log set around the next rise in memory so we can see what RoonServer and RoonAppliance are doing before the system gets tight.

Please follow the log collection steps, then upload the resulting logs with our secure uploader. Once that’s done, note the exact local time, date, and what the memory usage looked like when it starts climbing again, so we can line that up with the server logs. If anything changed recently on this machine, such as the latest update, a power event, or attached storage, please include that too.

Thank you, and let us know once the logs are up.

3

Hello, and thank you for the reply. The memory growth is linear from the start of the service. Currently it is at around 6GB, I will let it grow for a few days until memory is close to exhaustion, then collect logs and send them.

4

Perfect, thank you @Helge_Gudmundsen that sounds ideal.